Research & development - Leuven | More than two weeks ago
Respiratory-related diseases, such as chronic obstructive pulmonary disease and asthma, are one of the leading causes of death in the world. A reliable diagnosis is critical for proper treatment, medical management and prevention of an impending respiratory problem. In recent years, different types of digital stethoscopes have emerged in the market, replacing the conventional stethoscopes. A digital stethoscope converts an acoustic sound to electronic signals, which can be further processed to address respiratory problems.
The goal of this master’s thesis/internship is to develop a machine learning framework which can help in the automated detection and classification of respiratory problems, e.g. COPD. The student is expected to perform a study of the state-of-the-art literature, develop and verify a model on publicly available datasets. The developed algorithm should be implemented on a microcontroller platform (ARM Cortex) or FPGA to demonstrate its feasibility for real-time operations.
1) Literature review of existing algorithms, databases, systems and understand the basic physiology of respiratory diseases.
2) Algorithm development requiring time/frequency domain signal processing and machine learning (or deep learning).
3) Evaluate the model and benchmark results against state-of-the-art.
4) Implement the algorithm on ARM Cortex or FPGA to demonstrate feasibility of real-time operations.
Type of project: Internship, Thesis
Duration: 6 - 9 months
Required degree: Master of Engineering Technology, Master of Bioengineering, Master of Science, Master of Engineering Science
Required background: Biomedical engineering, Computer Science, Electrotechnics/Electrical Engineering
Supervising scientist(s): For further information or for application, please contact: Dwaipayan Biswas (Dwaipayan.Biswas@imec.be)
Imec allowance will be provided.